Running after 40, 50, 60: what actually changes
Endurance performance doesn't fall off a cliff at 40. Leyk et al.'s 2007 study of 300,757 marathon and half-marathon finishers found no significant age-related loss in finish time before age 50, then a decline of 2.6% to 4.4% per decade through the 50s and 60s. Tanaka and Seals's 2008 review describes the underlying pattern as modest decline to age 50 to 60, then steeper. Fleg et al.'s 2005 longitudinal study of 810 healthy adults found the rate of VO2max decline itself accelerates, from 3% to 6% per decade in the 20s and 30s to over 20% per decade in the 70s. What changes by decade and what to adjust are below.
Three peer-reviewed sources measured three different things: race finish times across a huge field, VO2max in a longitudinal cohort tracked for years, and a review synthesizing both. None of them measured the same population the same way, so the table below keeps each figure attached to what it actually came from.
What changes, by decade
This table is Runked's own, built from four sources: Leyk et al. 2007 (marathon and half-marathon finish times), Fleg et al. 2005 (longitudinal VO2max), Tanaka and Seals 2008 (a review of the physiological mechanisms), and Lepers, Burfoot and Stapley 2021 (a small longitudinal cohort of trained masters marathoners).
| Age band | Measured performance change | What's changing physiologically | Training adjustment a named source supports |
|---|---|---|---|
| 30s to ~50 | No significant age-related loss in marathon or half-marathon finish time, per Leyk et al. 2007 (n=300,757 finishers). Tanaka & Seals's 2008 review describes this range differently: peak endurance performance held only to about 35, with modest decline from there. | Fleg et al. 2005 (Baltimore Longitudinal Study of Aging, n=810, ages 21–87, free of clinical heart disease): VO2max decline is already underway and accelerating, starting near 3% to 6% per decade in the 20s and 30s. | Tanaka & Seals 2008: the review says declines in endurance performance and its physiological determinants with ageing appear to be mediated in large part by a reduction in the intensity and volume of exercise that can be performed in training. |
| 50s–60s | 2.6% to 4.4% decline per decade in finish time for the 50-to-69-year-olds in the same Leyk et al. 2007 dataset. | Tanaka & Seals 2008 describe this range as still "modest decreases," with the review naming reduced VO2max as the primary mechanism behind the loss. | Lepers, Burfoot & Stapley 2021 (n=40 sub-3-hour marathoners, tracked across 5 decades): those who kept racing and training consistently held their decline under 7% per decade, at least to age 60. |
| 70s and beyond | Fleg et al. 2005: VO2max decline reaches over 20% per decade by the 70s, up from 3% to 6% per decade in the 20s and 30s. The rate was larger in men than women from the 40s onward, per the same study. | Tanaka & Seals 2008 describe this range with the phrase "progressively steeper declines thereafter," referring to what follows the 50-to-60 plateau. | Lepers, Burfoot & Stapley 2021 note their own cohort's sub-7%-per-decade figure was only confirmed to age 60; the authors say further study is needed to see if it holds past that. |
The why, sourced
The two starkest figures on the table, Leyk's 2.6% to 4.4% per decade and Fleg's over 20% per decade past 70, aren't measuring the same thing. Leyk et al. 2007 timed actual race finishes in a field of 300,757 marathon and half-marathon runners, a self-selected group of people still racing at that age. Fleg et al. 2005 measured peak treadmill VO2max directly in 810 healthy, community-dwelling adults tracked longitudinally over a median 7.9 years, most of whom weren't competitive runners. VO2max decline is the mechanism Tanaka and Seals's 2008 review names as primary; a study of finish times in active racers, adjusted by nothing but the clock, will understate that decline compared with a lab measurement in a broader population, because the racers who keep entering marathons into their 60s and 70s are already a fitter-than-average slice of their age group.
Lepers & Cattagni's 2012 analysis of the best male and female finishers aged 20 to 79 at the New York City Marathon, 1980 to 2009, found finish times for men older than 64 and for women older than 44 improved significantly (P < 0.01) across those three calendar decades instead of plateauing, and concludes in its own closing section: "The present data suggest that male (≥65 years) and female (≥45 years) master runners have probably not yet reached their limits in marathon performances." That's a statement about the population's best performers getting faster over calendar time as training and participation grew. It isn't a claim that any individual runner's own decline reverses with age.
How to apply this
None of the sources above say the same rate of decline applies to everyone starting in their 40s, 50s or 60s. Lepers, Burfoot and Stapley's 2021 cohort held decline under 7% per decade, a group that had kept training and racing at a competitive volume throughout; the study is a retrospective analysis with no comparison arm, so it describes what that group achieved without establishing the training as the cause. Two concrete steps come out of the sourced sections above:
- Track your own trend against your own history. Leyk's 2.6% to 4.4% and Fleg's decade-by-decade acceleration are population averages across thousands of people; your own rate depends on the training you actually do. Watching your own logged runs slow or hold steady over months is a more direct signal of your own aging curve than any population table can give you. See how the running rank works.
- Keep the training volume up if you can tolerate it. Tanaka & Seals 2008 name training intensity and volume as the lever most tied to how much of the decline shows up. The strength training for runners guide covers one part of that: the CDC's own guidance for adults 65 and older calls for muscle-strengthening activity on 2 or more days a week alongside 150 minutes of moderate aerobic activity and balance work.
Anyone starting to run for the first time in their 40s, 50s or 60s should read the CDC's older-adult activity guidance first, whether or not they ran when younger: it recommends talking to a doctor before starting vigorous-intensity activity, including jogging, if you've been inactive, have a disability, are overweight, or have a chronic condition, and building toward the 150-minutes-a-week target gradually.
Common mistakes
- Comparing your finish time to a race field average. Leyk's sample is racers who were still fit enough to enter a marathon or half marathon at that age; someone starting later or training less won't match that curve.
- Assuming the same figure applies at every distance and intensity. Leyk's finish-time percentages and Fleg's VO2max percentages come from different measures on different populations; neither one converts directly into the other.
- Treating "modest decreases" as no decline at all. Tanaka and Seals's review uses that exact phrase for the 35-to-50/60 range, but it still describes ongoing loss, just at a slower rate than what follows it.
- Skipping a doctor check before jumping into vigorous training. The CDC's guidance is specific about who should check first: anyone who's been inactive, has a disability, is overweight, or has a chronic condition, before starting jogging or similar vigorous-intensity activity.
Talk to a doctor before starting a new running routine if you've been inactive, have a disability, are overweight, or have a chronic condition, per the CDC guidance above. Any symptoms during exercise that concern you are also worth a doctor's opinion.
What to do next
The age-graded running times guide covers the WMA/USATF tables, a separate age-and-sex-adjusted percentage system built for comparing a single race result against a published standard; it doesn't track decline over years the way the sources above do. Compare to yourself, not to others: a Runked rank starts everyone at Bronze and only moves on your own runs, so it's built for exactly the kind of year-over-year comparison this page is about.
Find out where you stand
Runked is free on the App Store. Connect your watch or track 3 runs, and meet your Runk.
Download on App StoreFrequently asked questions
Does running performance really decline before 50?
The two sources here disagree on when it starts, and both are stated as they measured it. Leyk et al.'s 2007 analysis of 300,757 marathon and half-marathon finishers found no significant age-related loss in finish times before age 50. Tanaka and Seals's 2008 review of the masters-athlete literature describes peak endurance performance as maintained only to about 35, with modest decreases from there to 50 to 60. One measured actual race times across a huge field; the other synthesized physiological studies. Neither is wrong; they measured different things.
Is it safe to start running for the first time at 50 or 60?
The CDC's older-adult activity guidance says to talk to your doctor before starting vigorous-intensity activity, including jogging, if you've been inactive, have a disability, are overweight, or have a chronic condition, then build toward 150 minutes a week of moderate aerobic activity (or 75 of vigorous), muscle-strengthening on 2 or more days a week, and balance work. This page is general guidance. It is not a personal medical clearance.
Does training slow the decline?
Lepers, Burfoot and Stapley's 2021 study followed 40 sub-3-hour marathoners across five decades and found that consistent training and racing limited their age-related decline in marathon performance to under 7% per decade, at least to age 60. That's a narrow group: 39 men and 1 woman, all lifelong competitive runners, so it shows what sustained training can achieve. It isn't necessarily typical for someone starting later in life. The authors say further study is needed to see if the low rate holds past 60.
Does Runked account for age when it scores my runs?
Not against a published age table. Runked PRO's race time predictions come from your own training history. There's no age-decline curve behind them, so they move with your own trend, whatever decade you're running in. Runked doesn't publish age-graded percentages; see the age grading guide for that separate, WMA-table-based approach.